We propose a program of four projects bearing on the phase problem of crystal structure determination. The immediate objective is to develop more powerful methods of biomolecular structure determination, first, by developing improved experimental and computational strategies for existing methods of sequential or incremental phasing and, second, by developing entirely new direct methods of global phasing. Success in these endeavors will render the phase problem routinely solvable for structures of several hundreds of atoms. The longer-range goal is progress toward development of direct methods that will be routinely applicable for de novo determinations of the structures of biological macromolecules composed of thousands of atoms. Eventually, this will obviate the need for the time-consuming and labor-intensive search for and chemical preparation and diffraction analysis of multiple isomorphous heavy-atom derivative crystals, which are required by presently used methods of macromolecular structure determination. Our program will involve the efforts of an exceptionally talented and experienced group of crystallographers at the Medical foundation of Buffalo. Members of this group have collaborated variously and successfully over a period of more than 15 years, but often these collaborations have been only incidental to the diverse individual research projects of several collaborators. The proposed program is intended to support the a coordinated team effort, which will benefit from the cross-fertilization of ideas bound to develop out of the rich diversity of crystallographic expertise of the scientific team. The component projects and principal scientists of the program are: (I) Direct Methods of Phase determination - H.A. Hauptman and C.M. Weeks, (II) New Algorithms for Intractable Direct Methods Problems - D.A. Langs and G.D. Smith, (III) Macromolecular Phase Problem: Experimental Approaches - G.T. DeTitta and R.H. Blessing, (IV) Direct Methods Phasing in Electron Crystallography - D.L. Dorset.